The temperature-regulated, flexible, glass-core calorimeter containing nitrogen-saturated water at 4 degrees C yields a dose- to-water from 4 MV x rays that is 1.006 times greater than that recorded by an ionization chamber. It remains unresolved as to whether this result means that both systems yield essentially the same, accurate dose, or whether a thermal defect in the water and one or more inaccurate parameters applied to the ionization measurements may be of the same magnitude and direction. The dosimetry of neutron and heavy charged-particle beams is currently established with calibrated, A-150-plastic ionization chambers or calorimeters made of the same material. Because of uncertainties in physical parameters, the calibrations of neutron beams with ionization chambers have uncertainties of 5.4-10.8% depending on energy. Calibrations with A-150 calorimeters suffer from a 4% thermal defect, and the conversion of dose-to-(A-150) to dose-to-muscle has an uncertainty of 2.6-9.7% due to the kerma factor ratio of muscle to A-150. Because the atomic composition of tissue-equivalent liquid (TEL) can be readily adjusted to match that of the tissue of interest, i.e., muscle, a TEL calorimeter modeled on the present water calorimeter has the potential for reducing the uncertainty of the dosimetry of these radiations. So as to improve the accuracy of radiation-therapy dosimetry, it is proposed that the following research and development be undertaken: 1. Measure the thermal defect of water and a variety of TEL's using proton beams in the range 1-40 MeV, and evaluate the effect of dissolved gases and radical scavengers. 2. Measure the specific heat of promising TEL's relative to water using precision thermometric and electronic instrumentation. 3. Study the effect of low operating temperatures on the performance of the TEL calorimeter where increased viscosity will reduce convection currents. 4. Using 4 MV x rays, compare the dose-to-muscle from the TEL calorimeter with that obtained from ionization-chamber dosimetry and from the water calorimeter. 5. Using 160 MeV protons, and subsequently high- energy neutrons, compare the dose-to-muscle from the TEL calorimeter with that obtained from an A-150 ionization chamber. 6. Expand the evaluation of the water calorimeter to include a wide range of x-ray and electron beams, and compare the results with the doses obtained with ionization chambers.